Download James Eldridge BC Fossorial Native Mammals

ARTICLE IN PRESS
B I O L O G I C A L C O N S E RVAT I O N
x x x ( 2 0 0 7 ) x x x –x x x
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/biocon
Reintroduction of fossorial native mammals
and potential impacts on ecosystem processes
in an Australian desert landscape
Alex I. James*, David J. Eldridge
School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052, Australia
A R T I C L E I N F O
A B S T R A C T
Article history:
Substantial range declines of the greater bilby (Macrotis lagotis) and the burrowing bettong
Received 1 March 2007
(Bettongia lesueur) are thought to have had dramatic effects on ecosystem processes in the
Received in revised form
Australian arid zone because of their impacts on surface soils. The reintroduction of bilbies
27 April 2007
and bettongs into a reserve in central Australia provided an opportunity to compare their
Accepted 30 April 2007
ecosystem impacts with those of two prevalent fossorial animals; the exotic European rabbit (Oryctolagus cuniculus) and the native sand goanna (Varanus gouldii). Bilbies and bettongs
dug deeper and wider pits, excavating significantly more soil than rabbits or goannas. Pit
Keywords:
coverage was four-times greater, and significantly more soil was excavated in the reserve
Ecosystem engineering
where bilbies, bettongs and goannas were enclosed together compared with a site outside
Biopedturbation
the reserve where rabbits and goannas co-occurred, or within the reserve where goannas
Macrotis lagotis
occurred alone. Goannas dug fewer holes outside the reserve than in either of the reserve
Bettongia lesueur
paddocks. Litter and viable seed were restricted almost exclusively to the pits, and soil from
Oryctolagus cuniculus
pits had higher levels of labile carbon than non-pit surface soils. Compared with surface
Varanus gouldii
soils, bilby, bettong and goanna pits contained relatively more labile carbon than rabbit
pits. The significantly greater soil excavation by the bilbies and bettongs, and higher concentrations of carbon in their pits, relative to rabbit and goanna pits, demonstrate that
these reintroduced fossorial mammals play important roles in the creation of fertile
patches in arid landscapes. The results suggest that the extirpation of Australia’s mammal
fauna has been accompanied by a loss of key ecosystem processes.
Crown Copyright 2007 Published by Elsevier Ltd. All rights reserved.
1.
Introduction
Australia has suffered the highest rate of recently recorded
mammal extinctions, with more than 30% of its mammalian
fauna now listed as extinct, endangered or vulnerable (Short
and Smith, 1994; Department of Environment and Water Resources, 2007). This has largely been attributed to two introduced predators; the red fox (Vulpes vulpes) and the feral cat
(Felis catus) (Smith and Quin, 1996; Johnson, 2006), as well as
habitat fragmentation, overgrazing, altered fire regimes and
the invasion of feral herbivores such as the European rabbit
(Oryctolagus cuniculus) (Morton, 1990; Short and Smith, 1994).
The extinctions have mainly been medium-sized mammals;
and throughout most of semi-arid and arid Australia south
of the tropics, all ground–dwelling mammals in the critical
weight range of 35 g–5.5 kg (Burbidge and McKenzie, 1989)
have disappeared (Johnson, 2006). Furthermore, a third of
the extant marsupial species have disappeared from more
than 50% of their original geographic range (Maxwell et al.,
1996). These losses, and the extensive range declines of the
* Corresponding author.
E-mail address: [email protected] (A.I. James).
0006-3207/$ - see front matter Crown Copyright 2007 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.biocon.2007.04.029
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
2
B I O L O G I C A L C O N S E RVAT I O N
medium-sized native mammals, have undoubtedly had significant effects on ecosystem structure and function.
Associated with the decline in native mammals has been
an eruption in the range of the European rabbit; a mediumsized feral herbivore (Strahan, 1995). The effects of rabbit
grazing, combined with extreme drought, can have disastrous
consequences for the health and diversity of arid zone vegetation (Wood, 1988; Leigh et al., 1989). For example, grazing
by rabbits is implicated in the decline in recruitment of many
long-lived woodland trees such as Acacia spp. (Lange and Graham, 1983). Rabbits are also a preferred prey species of feral
predators, leaving native mammals vulnerable to predation
during periodic crashes in rabbit numbers (Jarman, 1986;
Smith and Quin, 1996; Johnson, 2006).
Many large-scale conservation efforts within Australia
have therefore focussed on reintroducing species to areas
where low densities of feral predators and herbivores are
maintained by baiting programs or the erection of predatorproof fences (Short et al., 1992; Short and Turner, 2000; Moseby and Read, 2006). Reintroduction of threatened species is an
increasingly popular goal of conservation, and is seen as a
viable means for restoring population size and preventing
extinction, with the ultimate goal being the re-establishment
of wild populations of the reintroduced species (Griffiths
et al., 1996). Most reintroduction projects have focussed upon
the dynamics of reintroduced populations, the practical
methods involved and the subsequent success or failure of
reintroduction (Fischer and Lindenmayer, 2000; Seddon
et al., 2007).
Although the restoration of ecosystems may be a corollary
or even a goal of reintroduction, few studies have examined
the effects of reintroductions on ecosystem health (however,
see Hägglund and Sjöberg, 1999). Since species are being restored to their former range, an examination of the impacts
of reintroduction on ecosystem health not only provides a
measure of restoration success, but also gives valuable insights into how systems may have functioned prior to the loss
of species, and therefore, prior to degradation. The reintroduction of a species will likely alter ecosystems not only
through the consumption of resources and other trophic level
effects, but also through the modification of habitat. Organisms that modify, maintain, create or destroy structure within
the physical environment have been termed ‘ecosystem engineers’ (Lawton, 1994; Jones et al., 1997; Hastings et al., 2007).
Such modifications can affect energy flows and the availability of resources to other organisms, including positive feedback to the engineer itself (Jones et al., 1997; Day et al.,
2003).The modification of the abiotic environment by ecosystem engineers can dramatically alter water flows, nutrient
levels, micro-organisms, seed capture and habitat quality
(Decaens et al., 2002). Ecosystem engineers may also increase
species richness, diversity and productivity by creating
patches of habitat differing in resource availability, thus enabling organisms of different resource requirements to coexist (Day et al., 2003; Odling-Smee et al., 2003). Numerous
examples from terrestrial and aquatic ecosystems have demonstrated that the removal or addition of ‘keystone’ modifiers
from a system can drastically alter the surrounding environment (Gutierrez et al., 2003; Zhang et al., 2003).
x x x ( 2 0 0 7 ) x x x –x x x
The impact of ecosystem engineering is predicted to be
greater in harsh environments such as arid and semi-arid
areas, where production is limited more by resource flows
than by trophic interactions, and where amelioration of habitat is likely to increase the survival of organisms and extend
their distributions (Bertness and Callaway, 1994; Crain and
Bertness, 2006). Given that reintroductions may often be to
degraded areas where species and habitat have been lost,
the engineering effects of reintroduced species may be of
even greater significance. It has also been suggested that ecosystem engineers can aid restoration efforts by reducing the
threshold effort or human input required to restore a landscape to a desired state (Byers et al., 2006).
The greater bilby and burrowing bettong are two marsupials that have been reintroduced to a handful of fenced
reserves in Australia. Both species are within the critical
weight range (Burbidge and McKenzie, 1989), and were once
widely distributed across arid Australia (Morton, 1990; Southgate, 1990a). Greater bilbies now occupy only 20% of their former range and are restricted to a small isolated population in
south-western Queensland and low-density populations in
the Tanami Desert in the Northern Territory, and the Gibson
and Great Sandy Deserts in Western Australia (Southgate,
1990a). Burrowing bettongs were one of the most widely distributed mammals on the continent (Finlayson, 1958) and
are now restricted to two predator-free islands off the Western Australian coast (Strahan, 1995). Along with the sand
goanna (Varanus gouldii), which is widely distributed throughout the arid zone, these two vertebrates create small pits or
depressions in the soil surface while foraging (Southgate,
1990b; Whitford, 1998; Robley et al., 2001). Bilbies and bettongs are omnivorous, and dig for seeds, invertebrates, bulbs
and fungi (Southgate, 1990b; Gibson, 2001; Robley et al., 2001),
while goannas generally forage for invertebrates and small
reptiles (King and Green, 1979).
Pits constructed during foraging capture resources that flow
across the landscape (Alkon, 1999; Whitford and Kay, 1999;
Zhang et al., 2003), intercepting water flows and entrained organic matter and seeds (Gutterman and Herr, 1981; Reichman,
1984; Boeken et al., 1995), and creating areas of higher moisture
and nutrient-rich hotspots of litter decomposition (Steinberger
and Whitford, 1983; Hawkins, 1996; Garkaklis et al., 2003;
Eldridge and Mensinga, 2007). The creation of these ‘fertile
patches’ may be an important process that enhances the germination and establishment of plants in resource-depleted
landscapes. Pits contribute to microsite-level patchiness, leading to landscape-level increases in spatial heterogeneity of
resources. For example, the foraging pits of goannas structure
mulga (Acacia aneura) landscapes by trapping litter and seeds
on the upslope edge of the timbered mulga groves, thus
maintaining zonation between the resource-rich groves and
the resource-poor inter-groves (Whitford, 1998).
This study was conducted at Arid Recovery, a feral animalproof exclosure in northern South Australia. The presence of
reintroduced bilbies, bettongs and goannas in a large reserve,
as well as populations of rabbits and goannas outside the
reserve, provided the opportunity to compare the effects of
foraging pits created by three native vertebrates with that of
pits created by the exotic rabbit. Specifically, the objective
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
B I O L O G I C A L C O N S E RVAT I O N
was to examine the engineering effects of the four vertebrate species on soil chemistry, resource capture and plant
germination, and therefore, the implications for ecosystem
processes in arid shrubland.
The aim of this study was to examine the magnitude of
soil excavation and resource capture at both microsite (individual pits) and landscape scales within three landscape elements; dunes, swales and the intervening ecotones, and
within three paddocks occupied by different complements
of animals that all excavate pits. This study tested a number
of predictions about litter and seed capture and soil carbon in
relation to foraging pits and landscape elements (dune, ecotone, swale), and about the density of foraging pits in relation
to species composition and landscape element. Specifically,
these predictions were that: (1) all ecosystem engineers would
excavate more soil and dig more pits in the ecotones, because
these sites contain elements of both dunes and swales and
support the most productive and diverse vegetation community, (2) resource (litter, seed) capture would be greatest in the
pits, where litter, soil and seed is retained in situ rather then
being redistributed by wind and water, (3) the effect of pits
and subsequent litter capture on the concentration of labile
carbon would be greatest in the swales, where clay contents
are greatest, and least on the dunes, and, (4) the ecosystem
response would be highly species-dependent, in particular,
the effects of bilbies and bettongs would be greater than that
of rabbits. The purpose of testing these predictions was to
determine whether the loss of reintroduced fossorial animals
alters ecosystem processes and if so, in which part of the
landscape the effects are likely to be greatest.
2.
Methods
2.1.
The study site
Arid Recovery is located 20 km north of Roxby Downs in arid
South Australia (3029 0 S, 13653 0 E). The landscape is characterised by linear, east-west trending sand dunes and inter-dunal
swales, often with a variable surface cover of stones (‘gibbers’)
up to 5 cm in diameter. Dunes were about 200 m wide, had
sandy topsoils (5–10% clay) and supported an open shrubland
of sandhill wattle (Acacia ligulata) and narrow-leaved hopbush
(Dodonaea viscosa). The intervening gibber swales were about
500 m wide, had sandy-clay topsoils (35–40% clay) and were
dominated by shrubs from the family Chenopodiaceae (Atriplex vesicaria, Maireana astrotricha). The ecotones, intermediate
areas between the swales and dunes, were 10–20 m wide,
characterised by fine sandy topsoils (25% clay), and vegetated
by species found in both the dunes and swales.
The climate is arid, and the rainfall temporally and spatially variable, failing to reach the long-term average of
166 mm in 60% of years (Moseby and O’Donnell, 2003). At
the time of the study in December 2003, the mean daily maximum temperature was 34.5 C and the mean daily minimum
was 19.3 C (Bureau of Meteorology, 2004). The mean annual
maximum temperature exceeds 35 C, and the mean annual
minimum is 4 C (Olympic Dam Operations, 1994).
The Arid Recovery reserve is an 86 km2 exclosure in which
cats, foxes and rabbits have been eradicated and are pre-
x x x ( 2 0 0 7 ) x x x –x x x
3
vented from reinvading by a 2.5 m electrified vermin-proof
fence. Locally extinct mammals were reintroduced into the
exclosure in 1999. The exclosure is unique; its large size
means that it accommodates the home ranges of the reintroduced animals. Apart from the Perron Peninsula exclosure in
Western Australia, Arid Recovery is the largest area of intact
arid zone shrubland into which locally extinct animals have
been reintroduced.
The study was conducted at two sites within, and one site
outside, the reserve. The first reserve site was within ‘Main
Exclosure’, a 14 km2 paddock where the locally extinct burrowing bettong, greater bilby, greater stick-nest rat (Leporillus
conditor) and western barred bandicoot (Perameles bougainville)
have been reintroduced. As the latter two species do not forage in the soil, they were not included in the current study.
The second reserve site was within ‘Second Expansion’, an
8 km2paddock within which no locally extinct species have
been introduced. However, both Main Exclosure and Second
Expansion contain populations of sand goannas. The third
site was outside the reserve (termed ‘Outside’) within a mining lease operated by Olympic Dam Operations. The general
area is grazed intermittently by cattle at relatively low stocking rates. However, there has been no cattle grazing in the
Outside paddock since the reserve was established, though
it is still grazed by rabbits and contains goannas. The three
paddocks were therefore relatively similar in their vegetation
structure and composition, and would represent typical,
though slightly degraded, landscapes within which these animals would have occurred prior to European settlement.
The estimated density of rabbits at the time of the study
was 40 per km2, an average figure for the four years prior to
the study following the release of rabbit hemorrhagic disease
(RHD) in 1996, which substantially reduced population densities. However, rabbit densities did fluctuate between
10 per km2 and 80 per km2 in the previous four years in response to rainfall, extended dry periods and outbreaks of
RHD (Arid Recovery, 2005). Although the bilbies and bettongs
were enclosed within Main Exclosure, their estimated densities of 14 per km2 for each species (Arid Recovery, 2005) were
within the range of estimated natural population densities
ranging from more than 70 bettongs per km2 (Noble, 1995)
to eight bilbies per km2 made by Le Soeuf and Burrell on the
Nullarbor Plain in 1921 (Southgate, 1990a). Goannas are able
to move freely between the three paddocks as juveniles. Estimates of the relative densities of goannas based on measurements of track density were made using 500 m track
transects. These indicated 109.6 ( ± 13.6; mean ± SEM (Main),
41.6 ± 4.6 (Second) and 18.0 ± 1.4 (Outside)) tracks per km (Arid
Recovery, unpublished data).
2.2.
Experimental design
Three ‘blocks’ were selected within each of the three paddocks, separated by distances of about 750 m. Each block consisted of the three landscape elements; dune, swale and
ecotone. Measurement sites were randomly selected from
within each landscape element. Main Exclosure contained
bilbies, bettongs, and sand goannas, Second Expansion goannas only, and Outside, rabbits and goannas. This design
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
4
B I O L O G I C A L C O N S E RVAT I O N
enabled comparisons to be made between: (1) the three complements of species occurring in the different paddocks, (2)
goannas either on their own or with different animals, and
(3) bilbies/bettongs and rabbits. In terms of the effects of individual animal species, the study was not orthogonal i.e. not
all animals were found in all three paddocks. Because there
is only one Arid Recovery reserve, and it was not possible to
replicate the treatments elsewhere, the design is therefore
pseudoreplicated, and does not allow generalisation to be
made about the effects of ecosystem engineers beyond the
study site. Nevertheless, a single replicate of a unique ecosystem represents a valuable opportunity to gain information
about the effects of both locally extinct and feral species.
2.3.
Pit density and soil excavation
Two-metre wide belt transects of variable length (50–250 m,
depending on the density of pits), and aligned parallel to the
direction of the dunes, were used to assess the density of foraging pits of each species at the 27 sites. Only foraging pits
were measured, i.e. burrow systems of any animals were
not included, and only three burrows were encountered
across the 27 sites. The depth, length and width through the
centre of each pit encountered on the belt transect were measured, and the type of animal responsible for its creation was
determined, based on shape, depth and excavation angle. Pits
constructed by bilbies and bettongs were pooled because they
could not be distinguished easily from each other due to similar morphologies. While most pits could be assigned to a particular animal, especially those Outside and in Second
Expansion, a small proportion of shallow pits in Main Exclosures could have been created by different animals. For the
paddock and landscape-scale calculations, we used all pits,
irrespective of the engineer creating them. However, for studies of animal effects, only pits of a known origin were used.
The length, width and depth of an additional 94 pits of a
range of sizes and shapes constructed by rabbits, bilby/bettongs and goannas were used to calculate the mass of soil
excavated by animals from measurements made of individual
pits. Their volume was then calculated by measuring the volume of fine sand required to fill them using a volumetric cylinder. Volume was converted to mass by adjusting for bulk
density using soil bulk density data measured in triplicate
for each of dune, ecotone and swale soil. A range of linear
and non-linear models was fitted to the relationship between
soil mass and pit volume, and regression analyses used to
evaluate those models with the greatest explanatory power.
Irrespective of the type of pit or the animal that created it,
the simple product of pit length, width and depth explained
92% of the variance in soil mass (F1,92 = 935, P < 0.0001) and
was used to convert pit volume to soil mass for all holes measured at all sites.
2.4.
Litter capture and soil nutrients in pits
To assess whether pits trapped litter, and if larger pits trapped
more litter, the contents of five randomly selected pits were
collected, along with a sample for an adjacent non-pit surfaces of the same area, at each of the nine sites in each of
the three paddocks (n = 270 samples). Samples were dried at
x x x ( 2 0 0 7 ) x x x –x x x
40 C for 24 h and weighed. To examine whether nutrient concentrations of pit soils were higher than non-pit soils, samples of the uppermost 2 cm of soil were collected at the
same 270 pit and paired non-pit locations, air-dried and
passed through a 2 mm sieve. The labile carbon content of
these soils was measured using a simplified laboratory technique whereby slightly alkaline, dilute KMnO4 reacts with
the readily oxidisable (labile) carbon, converting Mn(VII) to
Mn(II), and lowering the absorbance of 550 nm light (Weil
et al., 2003).
2.5.
Germination of seeds contained in soil and litter
To determine if there was more viable seed collected in the
pits than the surface soil, litter and the underlying 2 cm of soil
were collected from an additional 10 pits and 10 adjacent
non-pits at each of the nine sites in Main Exclosure only
(n = 180 samples). Approximately, 50 g of soil from pits and
non-pits was scattered over a layer of approximately 2 kg of
propagation sand in shallow trays measuring 173 · 142 ·
55 mm. Control trays were also set up containing propagation
sand only to control for the presence of any glasshouse
weeds. Trays were placed in the glasshouse at average temperatures ranging from 14 C to 19 C, and allowed to germinate under natural light conditions. An automatic sprinkler
system delivered water to the trays for 1 min twice daily
(09:00 and 15:00 h). The trial was run for nine weeks from
March 16th to May 17th and seedlings counted when they
emerged from the soil. On July 16th, the litter samples described above were placed in the glasshouse on new propagation sand and the pots watered until October 8th 2004.
Seedlings were counted and removed from the trays once
they could be identified. Unidentified seedlings were transplanted to larger pots and grown until they could be positively
identified.
2.6.
Statistical analyses
For all tests, the basic unit of measurement was the site.
Therefore, individual pit data (e.g. pit dimensions, soil and litter capture) were averaged across transects at each site before
statistical analyses. Differences in pit density (loge) and soil
excavation (loge) were tested using a mixed-models ANOVA
after checking for homogeneity of variance using diagnostic
tools (e.g. normal and residual plots) within the Genstat statistical package (Payne et al., 1993). The mixed-models ANOVA had two error terms; a whole-plot stratum, which
considered paddocks i.e. different combinations of animals;
goannas with bilby–bettongs (Main), goannas alone (Second),
goannas with rabbits (Outside), and a sub-plot stratum, which
considered landscape elements (dune, gibber plain, ecotone)
and its interaction with paddocks.
For labile carbon, differences in relation to paddock, landscape element and microsite (i.e. pit vs. non-pit), and their
interactions were examined using a mixed-models ANOVA.
The whole-plot and sub-plot strata were the same as above.
A third stratum examined pit effects and the two- and
three-way interactions with paddocks and landscape element, and a fourth stratum accounted for variance among
the pits and surface samples for each landscape ele-
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
B I O L O G I C A L C O N S E RVAT I O N
ment · paddock combination. The same approach was used
to test for differences in litter capture (in pit samples only)
in relation to paddock and landscape element. The analysis
was restricted to pit samples because litter was found at only
four of the 135 surface locations, and the paucity of data precluded any formalized statistical testing for litter mass. The
whole- and sub-plot strata were the same as above, while a
third stratum accounted for the variance among the 10 pits
for each landscape element · paddock (animal treatment)
combination. For all analyses, significant differences between
means were examined using least significant difference
testing.
3.
Results
3.1.
Soil excavation and density of foraging pits
Averaged across all paddocks, pits occupied 0.9% of the surface of the dunes and ecotones, but only 0.2% of the surface
of the swales. Pit density also varied between landscape elements, with fewer pits in the swales (1515 ± 342 pits ha1,
mean ± SEM) compared with the dunes or ecotones (4667 ±
968 pits ha1; F2,12 = 8.87, P = 0.004, Fig. 1a), but there was
no significant landscape element by paddock interaction
(F4,12 = 0.51, P = 0.727). Trends in soil excavation were similar
x x x ( 2 0 0 7 ) x x x –x x x
to pit density, with significant differences between landscape
elements (F2,12 = 26.0, P < 0.0001). The mass of excavated soil
averaged 5.99 ± 1.55 t ha1 in the dunes, 5.06 ± 1.55 t ha1 in
the ecotones and 1.27 ± 0.33 in the swales (Fig. 1b). Ninety percent of soil excavation was on the dunes and ecotones.
In general, pits excavated by bilby–bettongs and goannas
were deeper (median depth = 80 mm) than rabbit pits
(50 mm), and the openings of bilby–bettong foraging pits were
wider (140 mm diameter) than either goanna (110 mm) or rabbit (90 mm) pits. There were substantial differences between
the three different groups of engineers. On average, pit coverage was four-times greater where bilbies, bettongs and goannas occurred together (Main, 4.3%) compared with sites where
rabbits and goannas occurred together (Outside, 1.2%), and
very low where goannas occurred alone (Second, 0.6%). Pit
density also differed significantly between the three groups
of animals (F2,6 = 31.47, P = 0.001), ranging from 6813 ±
1500 pits ha1 for bilbies–bettongs plus goannas (Main), to
2878 ± 585 pits ha1 (rabbits and goannas, Outside) and
1157 ± 260 pits ha1 for goannas only (Second; Fig. 1a). More
soil was excavated in Main (bilby–bettongs with goannas)
compared with Second (goannas alone) or Outside (goannas
with rabbits; F2,6 = 13.30, P = 0.012, Fig. 1b). There was no significant interaction, however, between species and landscape
element (F4,12 = 3.18, P = 0.699).
Goannas dug fewer holes when they occurred with rabbits
than when they occurred either on their own or with bilbies
and bettongs (F2,6 = 5.66, P = 0.042; Fig. 2a), and this was consistent across all three landscape elements. Goannas excavated significantly more pits in the dunes and ecotones
compared with the swales (F2,12 = 7.79, P < 0.009), but there
was no significant paddock by landscape element interaction
(F4,12 = 0.46, P = 0.76). The mass of soil excavated by goannas
in the presence of rabbits was only about a third that in the
presence of bilbies and bettongs, or when they occurred alone
(F2,6 = 5.6, P = 0.043, on log10x + 0.01 transformed data; Fig. 2b).
Trends among landscape elements were the same as for soil
excavation. Goannas dug similar-sized pits (mass soil pit1)
when they occurred alone, with rabbits, or with bilbies and
bettongs (F2,6 = 0.66, P = 0.55).
While bilbies and bettongs dug about four-times as many
pits in Main as rabbits dug outside the reserve (Fig. 2a), the results were not significant because of the variability in density
between sites (P = 0.21). However, bilbies and bettongs excavated significantly more soil (five to eight-times) than rabbits
(F1,4 = 15.7, P = 0.017; Fig. 2b), and soil excavation was significantly greater in the dunes and ecotones compared with the
swales (F1,4 = 15.7, P = 0.017).
3.2.
Fig. 1 – Mean (a) density of pits (pits ha1) and (b) soil
excavation (t ha1) for all animals in relation to landscape
element. Bars indicate the 5% least significant difference for
different paddocks (P) and landscape elements (L).
Significant differences between the three paddocks
(M = Main Exclosure, S = Second Expansion, O = Outside),
summed over landscape elements, are indicated.
5
Foraging pits, litter and soil carbon
Litter was found almost exclusively in the pits (16.1 ± 1.3 g)
compared with the surface immediately adjacent to the pit
(0.07 ± 0.04 g), and this was consistent across all paddocks
and landscape elements. Litter was found in only four surface
samples. Pits in the ecotones and swales captured a greater
mass of litter (18.0–20.4 g) compared with pits in the dunes
(9.9 g; F2,12 = 4.33, P = 0.038). Concentrations of labile carbon
(sqrtx+0.5 transformed) were greatest in the pits compared
with the surface (F1,18 = 142.0, P < 0.0001) and greatest in the
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
6
B I O L O G I C A L C O N S E RVAT I O N
x x x ( 2 0 0 7 ) x x x –x x x
engineer in one landscape. Overall, the increase in labile carbon with increasing litter mass was greatest for bilbies and
bettongs and least for rabbits.
3.3.
Foraging pits and germination
No plants germinated from surface soil, and only four Digitaria ciliaris (Poaceae) plants emerged from soil taken from
the pits. However, 1307 individuals from 46 genera emerged
from litter samples taken from the pits. Three species, two
forbs from the family Chenopodiaceae (Atriplex holocarpa, A.
vesicaria) and the grass Eragrostis dielsii, contributed 53% of
all germinants (A. James, unpublished data).
4.
Fig. 2 – Mean (a) density of goanna pits (pits ha1) and (b) soil
excavated by goannas (t ha1) in relation to landscape
element. Bars indicate the 5% least significant difference for
different paddocks (P) and landscape elements (L).
Significant differences between the three paddocks
(M = Main Exclosure, S = Second Expansion, O = Outside),
summed over landscape elements, are indicated. Note that
only data from pits that could be positively ascribed to
goannas are presented here.
swales (213 mg C kg1 soil), intermediate in the ecotones
(135 mg C kg1 soil), and least in the dunes (93.3 mg C kg1
soil; F2,12 = 19.68, P < 0.001; Table 1). Concentrations of labile
carbon in pits constructed by bilby–bettongs and goannas
were 50% greater than surface concentrations, while concentrations in pits constructed by rabbits were only 19% greater
(interaction: F2,18 = 12.1, P < 0.001). In general, there were poor
relationships between labile carbon concentration in the soil
and the mass of litter in the pits (R2 < 0.10), though relationships were stronger when we limited our analyses to one
Table 1 – Mean density of pits (pits ha1) and soil
excavation (t ha1) for bilbies–bettongs (Main Exclosure)
and rabbits (Outside)
Attribute
Dune
Ecotone
Swale
Bilby Rabbit Bilby Rabbit Bilby Rabbit
No. of pits ha1
Soil excavation
(t ha1)
1100a
4.29a
250b
0.43b
1017a
2.51a
150b
0.26b
186a
0.44a
83a
0.10a
For a given attribute, different superscripts indicate a significant
difference between bilbies–bettongs and rabbits for a particular
landscape element at P < 0.05.
Discussion
This study demonstrated that foraging pits constructed by
four ecosystem engineers contained more litter and seeds,
and were sites of enhanced labile carbon compared with the
intervening soil matrix. The two locally extinct, reintroduced
fossorial mammals (bilbies and bettongs) dug substantially
more pits than either goannas or rabbits, the remaining arid
zone engineers, suggesting that neither goannas nor rabbits
have assumed the engineering role of bilbies and bettongs. Indeed, the amount of soil excavated by rabbits was only 13–
20% that of bilbies and bettongs when the latter occurred at
population densities comparable to pre-European levels.
There is therefore strong evidence that the reintroduction of
locally extinct engineers is a critical issue to consider, not
only for species conservation, but for the creation of patches
of enhanced resources, and for the potential impacts on ecosystem function.
Foraging and subsequent pit formation created a mosaic of
two contrasting patch types; resource-rich pits, and the adjacent resource-poor soil matrix. In this study, the soil surface
had only a sparse litter cover, and foraging pits were effective
litter traps, typical of many desert environments (Reichman,
1984). Unlike surface soil, the soil in the pits was nutrientand seed-rich, and the germination experiment clearly supported the second prediction that resource capture is greatest
in the pits compared with the intervening soil matrix. Foraging pits and depressions are substantial repositories of seed
(Reichman, 1984), contrasting with the matrix where seed
densities are typically very low (Whitford, 2002). Indeed,
field-based observations after rainfall support the finding that
pits are foci for plant germination (Sparkes, 2001). The lowdensity of seed and litter in the matrix is attributed to surface
runoff (Shachak et al., 1991; Boeken et al., 1995), comminution
and photo-oxidation of litter by sandblasting (Moorhead and
Reynolds, 1989; Whitford, 2002), and the winnowing effect
of wind on litter and seed (Reichman, 1984). More litter was
trapped in the dunes and ecotones than in the swales, probably due to both the greater pool of available litter and the
movement of mobile sand into the pits, effectively trapping
litter in situ within successive layers of sand. The ability to retain nutrients will depend on the clay content of the soil, and
the substantially greater levels of clay in the swales (35–40%)
compared with the ecotones (25%) or dunes (5–10%) are driving the potential for different landscapes to retain mineral-
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
B I O L O G I C A L C O N S E RVAT I O N
ized nutrients, and thus significant differences in levels of
carbon found in each landscape.
This study demonstrated substantial differences in pit
density, soil excavation and litter capture between landscape
elements, with levels in the swales substantially lower than
in the dunes or ecotones, which were comparable. This partially confirms the first prediction that activity is greatest in
the ecotones, and suggests that pit-derived fertile patches will
not be created uniformly across all landscape elements, given
that formation depends on both the characteristics of the pits
(density, size, integrity, longevity) and the extant vegetation.
Variation in pit density at a site could be due to differences
in (1) digging rates, (2) pit longevity, or (3) a combination of
both (Alkon and Olsvig-Whittaker, 1989). In this study, the
variation lies with differences in digging rates, as those sites
with more pits corresponded with the landscape element in
which pits would have had a higher turnover (i.e. shorter
half-life), due to the mobility of the sandy soil that accumulates within them. While there was also significantly more
soil excavation in both the dunes and ecotones compared
with the swales (Fig. 2), the effect of foraging pits is predicted
to be greatest in the ecotones, where the soils have higher
nutrient levels than the dunes.
The ecotones receive litter and seed from both the dunes
and swales, while mobile sand from the dunes anchors litter
in the pits. Although the ecotone is a relatively narrow zone,
small differences in nutrients, seed retention and moisture
levels within the pits would be expected to increase the diversity of patch types, promoting the gemination and persistence
of both dune- and swale-dependent species. Irrespective of
landscape type, foraging pits, with their tendency to trap more
litter, and hence, more seeds would contribute to the development of enhanced sites of germination, consistent with studies in arid areas worldwide (Gutterman and Herr, 1981;
Reichman, 1984; Alkon and Olsvig-Whittaker, 1989; Shachak
et al., 1991; Boeken et al., 1995). The direct effect of animal digging and pit creation would therefore be to increase landscape-level productivity (Bianchi et al., 1989).
While mechanisms driving the formation of fertile patches
are similar for all animal species, differences in pit morphology and the magnitude of soil excavation will govern the degree to which different complements of engineers will
impact upon landscape processes. Few studies have considered multi-species engineering effects and the resulting landscape impacts (Brown and Heske, 1990; Milton et al., 1997;
Machicote et al., 2004), and this study indicates that, at least
for arid shrubland at Arid Recovery, all engineers are not
equal. Pits of bilbies and bettongs were deeper and wider than
those of rabbits, and the effectiveness of pits at enhancing soil
carbon over and above levels in the soil matrix was much
greater for bilby–bettong and goanna pits (50% increase) than
rabbit pits (19% increase). As predicted, there were significant
differences in soil excavation and pit density between species,
confirming the fourth prediction. The cover of pits within the
Main Exclosure was more than four-times greater than outside
the reserve. Together bilbies, bettongs and goannas dug significantly more pits (Main; Fig. 1) than goannas and rabbits together (Outside; Table 1), and both dug more pits and
excavated a greater mass of soil than goannas alone (Second;
Fig. 2).
x x x ( 2 0 0 7 ) x x x –x x x
7
Some differences in pit density could be attributed to
goannas, which had slightly higher densities in Main Exclosure compared with Second Expansion or Outside (Arid
Recovery, unpublished data). Goannas excavated significantly
more soil in Main Exclosure in the presence of the reintroduced species than they did in the Second Expansion where
they occurred alone. This is an interesting result, given that
goannas are free to move between exclosures as juveniles
and would not be adversely affected by the predator-proof
fencing. Indeed, outside the reserve, in the presence of rabbits, goannas excavated only one-third the amount of soil
as inside the reserve, although this difference can in part be
attributed to the presence of predators outside the fenced
reserve.
The critical issue arising from this study is whether the
European rabbit, the dominant fossorial mammal over much
of arid Australia, has assumed the ecosystem engineering
role of displaced bilbies and bettongs. The data demonstrate
that rabbits have not matched the mass of soil excavated by
locally extinct animals, and there is a suggestion that they
produce markedly fewer pits. Bilbies and bettongs dug about
four-times as many pits, accounting for five- to eight-times
more soil excavation than rabbits (Fig. 3) again confirming
the prediction of a species-specific engineering effect. This
effect is amplified when one considers per capita pit construction, given that there were approximately 40% fewer
bilbies and bettongs than rabbits. Densities of bilbies and
bettongs inside the Main Exclosure were well within the
range of reported natural pre-European densities (Southgate,
1990a; Noble, 1995). Furthermore, pit densities of bilbies and
bettongs in this study (200–1100 pits ha1) were slightly
below rates reported for other closely-related bettong species
such as Bettongia penicillata (5000–16,000 ha1 yr1, Garkaklis
et al., 2004), Bettongia gaimardi (500–3000 ha1, Johnson,
1994), and the potoriod Potorous tridactylus (2250 pits ha1
Fig. 3 – Labile carbon concentrations (mg C kg1 soil) of soil
from pits of three groups of animals and an adjacent soil
surface, across three landscapes. Bars indicate the 5% least
significant difference for different animal (A), landscape
element (L) and the landscape element by animal
interaction (L · A). Significant differences between the three
landscapes (D = dune, E = ecotone, S = swale), summed over
animal groups, are indicated.
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
8
B I O L O G I C A L C O N S E RVAT I O N
year1, Claridge et al., 1993). Although the relationship
between population size and pit density is likely to be complex, an understanding of per capita impacts is important, given historic fluctuation in rabbit densities and the fact that
rabbits are probably being suppressed by predators across
most of arid Australia.
Thus this study compared two scenarios, the current situation across arid Australia where rabbits and introduced predators exist in the absence of native fossorial mammals, and
the pre-European situation, where bilbies and bettongs were
widespread in the absence of rabbits and introduced predators. Regardless of their effect on fertile patch creation, it is
likely that any small contribution by rabbits would be greatly
outweighed by their devastating impacts on plant recruitment and survival in arid Australia (Lange and Graham,
1983; Denham and Auld, 2004). We argue therefore that the local extinction of native fossorial mammals is likely to have
had significant impacts on landscape processes, given that
their role has not been assumed by either existing native animals (goanna) or exotic herbivores (rabbits).
Reintroduction of bilbies and bettongs therefore likely has
positive and unique impacts on the restoration of fertile
microsites in arid Australia. The reintroduction of species
not only alters the number and distribution of fertile patches,
but also sheds light on some of the ecosystem engineering
processes that have been lost with the extirpation of Australia’s mammal fauna. The linkages between the main soil-foraging animals in arid Australia and their relative roles in
structuring and maintaining landscape heterogeneity in desert ecosystems is deserving of more attention.
Acknowledgements
We are extremely grateful to Arid Recovery for allowing us to
use their research facilities and study the effects of their animals. We thank Celeste Ellice, Jeff Turpin, Nicki Munro, Katherine Moseby and John Read for assistance with field work,
and Jess Mason, Sarah Howcroft and Jessica Bryant for assisting with laboratory work. Frank Hemmings identified many of
the glasshouse plants, and Geoff McDonnell helped with the
germination trials. We thank Ian Oliver, Jim Noble and two
anonymous reviewers for their insightful comments on the
manuscript, and Terry Koen who assiduously guided us
through the statistical issues. Finally, we could not have done
this work without Brydie Hill, who provided encouragement,
enthusiasm and logistical support whenever they were
needed.
R E F E R E N C E S
Alkon, P.U., 1999. Microhabitat to landscape impacts: crested
porcupine digs in the Negev Desert highlands. Journal of Arid
Environments 41, 183–202.
Alkon, P.U., Olsvig-Whittaker, L., 1989. Crested porcupine digs in
the Negev desert highlands: patterns of density, size, and
longevity. Journal of Arid Environments 17, 83–95.
Arid Recovery, 2005. Arid Recovery Annual Report, 2004. Arid
Recovery, Roxby Downs.
x x x ( 2 0 0 7 ) x x x –x x x
Bertness, M.D., Callaway, R., 1994. Positive interactions in
communities. Trends in Ecology and Evolution 9, 191–193.
Bianchi, T.S., Jones, C.G., Shachak, M., 1989. Positive feedback of
consumer population density on resource supply. Trends in
Ecology and Evolution 4, 234–238.
Boeken, B., Shachak, M., Gutterman, Y., Brand, S., 1995. Patchiness
and disturbance: plant community responses to porcupine
diggings in the central Negev. Ecography 18, 410–422.
Bureau of Meteorology, 2004. Climate Averages for Australian
sites: Averages for Andamooka [Webpage on internet]. The
Bureau of Meteorology, Canberra, ACT [updated 16.8.2004,
accessed 29.10.2004]. http://www.bom.gov.au/climate/
averages/tables/cw_016065.shtml.
Brown, J.H., Heske, E.J., 1990. Control of a desert–grassland
transition by a keystone rodent guild. Science 250, 1705–1707.
Burbidge, A.A., McKenzie, N.L., 1989. Patterns in the modern
decline of Western Australia’s vertebrate fauna: causes and
conservation implications. Biological Conservation 50, 143–
198.
Byers, J.E., Cuddington, K., Jones, C.G., Talley, T.S., Hastings, A.,
Lambrinos, J.G., 2006. Using ecosystem engineers to restore
ecological systems. Trends in Ecology and Evolution 21, 493–500.
Claridge, A.W., Tanton, M.T., Cunningham, R.B., 1993. Hypogeal
fungi in the diet of the long-nosed potoroo (Potorous tridactylus)
in mixed-species and regrowth eucalypt forest stands in
south-eastern Australia. Wildlife Research 20, 321–337.
Crain, C.M., Bertness, M.D., 2006. Ecosystem engineering across
environmental gradients: implications for conservation and
management. Bioscience 56, 211–218.
Day, R.L., Laland, K.N., Odling-Smee, J., 2003. Rethinking
adaptation: the niche construction perspective. Perspectives
and Biology and Medicine 46, 80–95.
Decaens, T., Asakawa, N., Galvis, J.H., Thomas, R.J., Amezquita, E.,
2002. Surface activity of soil ecosystem engineers and soil
structure in contrasting land use systems in Colombia.
European Journal of Soil Biology 38, 267–271.
Denham, A.J., Auld, T.D., 2004. Survival and recruitment of
seedlings and suckers of trees and shrubs of the Australian
arid zone following habitat management and the outbreak of
rabbit calicivirus disease (RCD). Austral Ecology 29, 585–599.
Department of Environment and Water Resources, 2007. Species
Profile and Threats Database: EPBC Act list of Threatened
Fauna [Webpage on internet]. Department of Environment and
Water resources, Canberra, ACT. [updated 30.01.2007, accessed
20.02.2007]. http://www.environment.gov.au/cgi-bin/sprat/
public/publicthreatenedlist.pl?wanted=fauna.
Eldridge, D.J., Mensinga, A., 2007. Foraging pits of the shortbeaked Echidna (Tachyglossus aculeatus) as small-scale patches
in a semi-arid Australian box woodland. Soil Biology and
Biochemistry 39, 1055–1065.
Finlayson, H.H., 1958. On central Australian mammals, part III:
the Potoroinae. Records of the South Australian Museum 13,
203–235.
Fischer, J., Lindenmayer, D.B., 2000. An assessment of the
published results of animal relocations. Biological
Conservation 96, 1–11.
Garkaklis, M.J., Bradley, J.S., Wooller, R.D., 2003. The relationship
between animal foraging and nutrient patchiness in
southwest Australian woodland soils. Australian Journal of
Soil Research 41, 665–673.
Garkaklis, M.J., Bradley, J.S., Wooller, R.D., 2004. Digging and soil
turnover by a mycophagous marsupial. Journal of Arid
Environments 56, 569–578.
Gibson, L.A., 2001. Seasonal changes in the diet, food availability
and food preference of the greater bilby (Macrotis lagotis) in
south-western Queensland. Wildlife Research 28, 121–134.
Griffiths, H.I., Davison, A., Birks, J., 1996. Species reintroductions.
Conservation Biology 10, 923.
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029
ARTICLE IN PRESS
B I O L O G I C A L C O N S E RVAT I O N
Gutierrez, J.L., Jones, C.G., Strayer, D.L., Iribarne, O.O., 2003.
Molluscs as ecosystem engineers: the role of shell production
in aquatic habitats. Oikos 101, 79–90.
Gutterman, Y., Herr, N., 1981. Influences of porcupine (Hystrix
indica) activity on the slopes of the Northern Negev
mountains-germination and vegetation renewal in different
geomorphological types and slope directions. Oecologia 51,
332–334.
Hägglund, Å., Sjöberg, G., 1999. Effects of beaver dams on the fish
fauna of forest streams. Forest Ecology and Management 115,
259–266.
Hastings, A., Byers, J.E., Crooks, J.A., Cuddington, K., Jones, C.G.,
Lambrinos, J.G., Talley, T.S., Wilson, W.G., 2007. Ecosystem
engineering in space and time. Ecology Letters 10, 153–164.
Hawkins, L.K., 1996. Burrows of kangaroo rats are hotspots for
desert soil fungi. Journal of Arid Environments 32, 239–249.
Jarman, P.J., 1986. The red fox: an exotic large predator. In:
Kitching, R.L. (Ed.), The Ecology of Exotic Animals and Plants:
Some Case Histories. Wiley, Brisbane, pp. 44–61.
Johnson, C.N., 1994. Distribution of feeding activity of the
Tasmanian bettong (Bettongia gaimardi) in relation to
vegetation patterns. Wildlife Research 21, 249–255.
Johnson, C., 2006. Australia’s Mammal Extinctions: A 50 000 Year
History. Cambridge University Press, Melbourne.
Jones, C.G., Lawton, J.H., Shachak, M., 1997. Positive and negative
effects of organisms as physical ecosystem engineers. Ecology
78, 1946–1957.
King, D., Green, B., 1979. Notes on diet and reproduction of the
Sand Goanna Varanus gouldii rosenbergi. Copeia 1, 64–70.
Lange, R.T., Graham, C.R., 1983. Rabbits and the failure of
regeneration in Australian arid zone Acacia. Australian
Journal of Ecology 8, 377–381.
Lawton, J.H., 1994. What do species do in ecosystems? Oikos 71,
367–374.
Leigh, J.H., Wood, D.H., Holgate, M.C., Slee, A., Stanger, M.G., 1989.
Effects of rabbit and kangaroo grazing on two semi-arid
grassland communities in central-western New South Wales.
Australian Journal of Botany 37, 375–396.
Machicote, M., Branch, L.C., Villarreal, D., 2004. Burrowing owls
and burrowing mammals: are ecosystem engineers
interchangeable as facilitators? Oikos 106, 527–535.
Maxwell, S., Burbidge, A.A., Morris, K., 1996. The 1996 Action Plan
for Australian Marsupials and Monotremes. Environment
Australia, Canberra.
Milton, S.J., Dean, W.R.J., Klotz, S., 1997. Effects of small-scale
animal disturbances on plant assemblages of set-aside land in
Central Germany. Journal of Vegetation Science 8, 45–54.
Moorhead, D.L., Reynolds, J.F., 1989. Mechanisms of surface litter
mass loss in the northern Chihuahuan desert: a
reinterpretation. Journal of Arid Environments 16, 157–163.
Morton, S.R., 1990. The impact of European settlement on the
vertebrate animals of arid Australia: a conceptual model. In:
Proceedings of the Ecological Society of Australia, vol. 16, pp.
201–213.
Moseby, K.E., O’Donnell, E., 2003. Reintroduction of the greater
bilby, Macrotis lagotis (Reid) (Marsupalia: Thylacomyidae), to
northern South Australia: survival, ecology and notes on
reintroduction protocols. Wildlife Research 30, 15–27.
Moseby, K.E., Read, J.L., 2006. The efficacy of feral cat, fox and
rabbit exclusion fence designs for threatened species
protection. Biological Conservation 127, 429–437.
Noble, J.C., 1995. Bettong (Bettongia spp.) biological control of
shrubs in semi-arid mulga (Acacia aneura) rangelands: an
hypothesis, In: Page, M.J., Beutel, T.S. (Eds.), Ecological
Research and Management in the Mulgalands. University of
Queensland, Gatton College, Gatton, pp. 139–145.
Odling-Smee, F.J., Laland, K.N., Feldman, M.W., 2003. Niche
construction: the neglected process in evolution. Monographs
x x x ( 2 0 0 7 ) x x x –x x x
9
in Population Biology. Princeton University Press, New Jersey.
37.
Olympic Dam Operations, 1994. Environmental Management
Program Annual Report 1994, Olympic Dam Operations,
Olympic Dam, South Australia.
Payne, R.W., Lane, P.W., Digby, P.G.N., Harding, S.A., Leech, P.K.,
Morgan, G.W., Todd, A.D., Thompson, R., Tunnicliffe Wilson,
G., Welham, S.J., White, R.P., 1993. Genstat 5 Release 3
Reference Manual. Oxford University Press, New York.
Reichman, O.J., 1984. Spatial and temporal variation of seed
distribution in Sonoran Desert soils. Journal of Biogeography 1,
1–11.
Robley, A.J., Short, J., Bradley, S., 2001. Dietary overlap between the
burrowing bettong and the European rabbit in semi-arid
coastal Western Australia. Wildlife Research 28, 341–349.
Seddon, P.J., Armstrong, D.P., Maloney, R.F., 2007. Developing the
science of reintroduction biology. Conservation Biology 21,
303–312.
Shachak, M., Brand, S., Gutterman, Y., 1991. Porcupine
disturbances and vegetation pattern along a resource gradient
in a desert. Oecologia 88, 141–147.
Short, J., Bradshaw, S.D., Giles, J., Prince, R.I.T., Wilson, G.R., 1992.
Reintroduction of macropods (Marsupialia: Macropodoidea) in
Australia – a review. Biological Conservation 62, 189–204.
Short, J., Smith, A., 1994. Mammal decline and recovery in
Australia. Journal of Mammalogy 75, 288–297.
Short, J., Turner, B., 2000. Reintroduction of the burrowing bettong
Bettongia lesueur (Marsupialia: Potoroidae) to mainland
Australia. Biological Conservation 96, 185–196.
Smith, A.P., Quin, D.G., 1996. Patterns and causes of extinction
and decline in Australian conilurine rodents. Biological
Conservation 77, 243–267.
Southgate, R., 1990a. Distribution and abundance of the greater
bilby Macrotis lagotis Reid (Marsupialia: Peramelidae). In:
Seebeck, J.H., Brown, P.R., Wallis, R.I., Kemper, C.M. (Eds.),
Bandicoots and Bilbies. Surrey Beatty, Sydney, pp. 293–302.
Southgate, R., 1990b. Habitats and diets of the greater bilby
Macrotis lagotis Reid (Marsupialia: Peramelidae). In: Seebeck,
J.H., Brown, P.R., Wallis, R.I., Kemper, C.M. (Eds.), Bandicoots
and Bilbies. Surrey Beatty, Sydney, pp. 303–309.
Sparkes, S., 2001. A comparative study on the impacts of greater
bilby Macrotis lagotis (Peramelidae) and the European rabbit
Oryctolagus cuniculus (Leporidae) digs on vegetation structure
in arid South Australia. Unpublished BSc (Hons.) thesis,
University of Adelaide.
Steinberger, Y., Whitford, W.G., 1983. The contribution of rodents
to decomposition processes in a desert ecosystem. Journal of
Arid Environments 6, 177–181.
Strahan, R., 1995. The Mammals of Australia. Reed, Sydney.
Weil, R.R., Islam, K.R., Stine, M.A., Gruver, J.B., Samson-Liebig, S.E.,
2003. Estimating active carbon for soil quality assessment: a
simplified method for laboratory and field use. American
Journal of Alternative Agriculture 18, 3–17.
Whitford, W.G., 1998. Contribution of pits dug by goannas
(Varanus gouldii) to the dynamics of banded mulga landscapes
in eastern Australia. Journal of Arid Environments 40, 453–457.
Whitford, W.G., 2002. Ecology of Desert Systems. Academic Press,
London.
Whitford, W.G., Kay, F.R., 1999. Biopedturbation by mammals in
deserts: a review. Journal of Arid Environments 41, 203–230.
Wood, D.H., 1988. The rabbit (Oryctolagus cuniculus L.) as an
element in the arid biome of Australia, In: Cogger, H.G.,
Cameron, E.E. (Eds.), Arid Australia. Australian Museum,
Sydney, pp. 273–287.
Zhang, Y., Zhang, Z., Liu, J., 2003. Burrowing rodents as ecosystem
engineers: the ecology and management of plateau zokors
Myospalax fontanierii in alpine meadow ecosystems on the
Tibetan Plateau. Mammal Review 33, 284–294.
Please cite this article in press as: James, A.I., Eldridge, D.J., Reintroduction of fossorial native mammals ..., Biol. Conserv.
(2007), doi:10.1016/j.biocon.2007.04.029